The present invention relates to the field of computers, and particularly to a method and apparatus for adding and processing tags with emotion data.
As Internet technologies constantly develop, users access various content such as pictures, text, video, audio and etc. Therefore, quickly and accurately searching and accessing desired content becomes more of an issue.
Searching content can be performed by automatic content analysis. However, automatic content analysis does not adequately generate sufficient content metadata because there is a semantic gap between low-level features and high-level concepts. Therefore, current technology is to add semantic tags for the content. By doing so, a user can generate tags proactively to describe the content in a more accurate way.
However, at present, all tags supported by various Web 2.0 services are only textual tags including keywords of text features, but such textual tags cannot always express special content sufficiently. For instance, a textual tag of a snow mountain picture includes the following keywords of text features: the Himalayas, travel, mountain, snow and so on, but this cannot express a user's emotions and cannot display the emotions of the user who sees the Himalayas for the first time.